ML17199U392
| ML17199U392 | |
| Person / Time | |
|---|---|
| Site: | Dresden, Quad Cities, 05000000 |
| Issue date: | 02/19/1988 |
| From: | Silady J COMMONWEALTH EDISON CO. |
| To: | Murley T Office of Nuclear Reactor Regulation |
| References | |
| 4208K, NUDOCS 8802290292 | |
| Download: ML17199U392 (15) | |
Text
.. *. ~.
i~
I Commonm:alth Edison One First Na~
Plaza, Chicago, Illinois Address Reply to: Post Office Box 767 Chicago, Illinois 60690 - 0767 Mr. Thomas E. Murley, Director Office of Nuclear Reactor Regulation U.S. Nuclear Regulatory Commission Washington, DC 20555 February 19, 1988
Subject:
Quad Cities Station Units 1 and 2 Dresden station Units 2 and 3 Effect of Postulated DC Power Failure NRC Docket Nos. 50-254/265 and 50-237/249 References (1):
December 21, 1987 letter from J.A. Silady to T.E. Murley (2):
December 22, 1987 letter from J.A. Silady to T.E. Murley (3):
December 22, 1987 letter from G.M. Holahan to L.D. Butterfield (4):
Januqry 21, 1988 letter from J.A. Silady to T.E. Murley (5):
February 5, 1988 Conference Call between CECo (N.J. Kalivianakis, et. al.,) and NRR (G.M.
Holahan, et. al.,) and RIII (M. Ring).
- 1
Dear M.r. Murley:
- ... I'*.
c ** M~].
In December 1987 Commonwealth Edison determined that our Dresden and
~g~' Quad Cities units were susceptible to a DC power failure scenario which was 1gg *,
first identified at the Fermi 2 plant.
The postulated scenario is an extremely
\\
- min*
iow probability sequence of events which could affect some BWR 3' s and BWR 4 's
- N* ~
1
- ~hich have the LPCI Loop Select Logic feature for mitigating Loss of coolant u-=:
~ccidents (LOCA).
The required sequence of events involves the loss of 125V NO*~ DC control power within 15 seconds of the postulated design basis LOCA assuming
~~ :; simultaneous Loss of Off site Power (LOOP).
I ru~
I IN 1
- oa: *.
References (l)' and (2) notified the NRC.. of this scenario~s QlQ.l.
1ma.o..: applicability to Dresden and Quad Cities, addressed its safety" significance,
\\
~nd presented an action plan to resolve the issue.
Reference (3) acknowledged
,. _) the CECo commitment and concurred that there should be no impact on plant
-operations while the issue is resolved, due to the minimal safety significance.
((EG-F!l£.S Pwt
T.E. Murley 2 -
February 19, 1988 The status of CECo efforts was presented during the Reference (5) conference call. This letter responds to the Reference (3) request for either a) additional justification for not considering DC power failures as part of the design base~ or b) proposed ~edifications which assure acceptable ECCS performance should a DC failure occur simultaneous with LOCA and LCX>P.
CECo believes that Sections A *:and 8 *of this letter present an adequate justificati9n for continued ex¢lusion of the DC power failures from the Dresden and *Quad Cities design bases.**.. *However, as discussed below, CECo also believes. that the existing 125V DC swing bus arrangement represents a design weakness which should be corrected even if not required by the existing design basis*.. section c theref°ore pr.ovides the CECo plans with respect to finalizing and implementing a design modification to correct this weakness.
Section D discusses the proposed revisions to the Quad Cities LOCA Analysis Report.
A.
PROBABILITY *AND SAFETY SIGNIFICANCE CECo qua11tatively evaluated in Reference (2) the likelihood of the postulated simultaneous occurrence of these three events and considered mitigating factors such as existing procedures for monitoring and surveillance of the station batteries. It was concluded that the postulated scenario is of minimal safety significance due to:
a)
The extremely unlikely sequence of events which must occur in a very narrow time period.
b)
The extremely high reliability of the battery system.
c) Regular surveillances which would detect even partial battery degradation.
d)
Continuous monitoring of key battery-related parameters via control room alarms.
e)
Interim procedures implemented to mitigate the concern.
The qualitative discussion of event probabilities has subsequently been supplemented by a) reviewing a General Electric probability assessment performed in 1979 for the Monticello Plant, based on a postulated DC power failure coincident with LCX>P and a small break LOCA, and b) evaluating the specific design of Quad Cities unit 1 with respect to probabilities of DC failures, LOCA and LCX>P.
T.t:. Murley 3
February 19, 1988 Attachment A discusses these reviews and concludes that the postulated sequence (i.e., coincident LOCA, LOOP and Battery Failure) has a probability on th~ order of 10-10 per year, which is well below the commonly accepted threshold. for safety significant sequences c-10-7) and the typical plant core damage frequency c-10-4).
B**
SINGLE FAILURE LICENSING BASIS In addition to developing more quantitative probability estimates for the postulated event sequence, CECo has continued to research the licensing basis history of the ECCS single failure assumptions in previous QUad Cities and Dresden LOCA analyses of record.
This additional review has confirmed the conclusions of our earlier evaluation* that all previously approved Quad Cities LOCA analyses have only considered single active component failures.
The Attachment B discussion has been confirmed with General Electric and is consistent with their under-standing of the Quad Cities ECCS analysis single failure assumptions.
The additional review also confirmed that Dresden LOCA analyses which have been performed by Advanced Nuclear Fl.J.els (formerly Exxon Nuclear) have similarly addressed active component failures only.
While Dresden and Quad Cities previous analyses of record have not addressed failures of passive components such as the 125V DC battery, several licensing documents have been identified which include discussions of the impact of DC power failure on the availability of remaining ECCS equipment.
Reference (4) identified a related Dresden FSAR Question and Answer and several other documents were discussed during the Reference (5) conference call. A list of these documents is also provided in Attachment B.
The conclusion of the CECo and General Electric reviews is that the assumption of a passive electric failure has not been a consistently applied ECCS analysis requirement, especially on older vintage plants such as Dresden and Quad Cities.
Neither 10 CFR 50.46 nor 10 CFR 50 Appendix K explicitly defines what constitutes a "single failure.
In 1971, Appendix A was added to 10 CFR 50 including a general definition of "single failure".
Although what constitutes a passive electrical failure was not explicitly defined, a footnote to that definition indicated that passive electrical failures should be considered.
In contrast, various revisions of the Staff's Standard Review Plan have subsequently been issued which indicate that only single active failures must be considered in the short term phase of a LOCA.
This is consistent with the Industry Standard (ANSI N658-1976) for single failure treatment in fluid systems.
Relevant excerpts from Section 6.3 of NUREG-0800 (Rev. 2 - April 1984) are also enclosed with Attachment B for your convenience.
- Originally provided as the Attachme.nt to Reference ( 1) and enclosed here in Attachment B for convenience. *.
- r
-J
- ,..J t *'
\\;
(.
- -j
... ~
- .-~
'r.::
.. 2,,.
T. February 19, 1988 Although the single failure licensing requirements are not well defined especially for older vintage plants, Commonwealth Edison believes that the previous Dresden and Quad Cities LOCA analyses, which have been NRC reviewed and approved, have clearly stated the basis for assumed single failures (as discussed in Attachment B) and have been in compliance with 10 CFR 50.46 and Appendix K.
C.
DESIGN MODIFICATION Although the postulated accident sequence is extremely remote and has not previously been considered in ECCS analyses performed in compliance with 10 CFR 50.46 and Appendix K, Commonwealth Edison acknowledged in Reference (2) that the 125V DC swing bus arrangement as currently configured represents a design weakness which should be corrected.
A feasibility study has therefore been completed and conceptual designs developed for both Quad Cities and Dresden which will eliminate the susceptability of these plants to the postulated failure scenario.
Attachment C describes the proposed modifications and Attachment D provides marked-up copies of the current electrical drawings as requested by your staff in the Reference (5) discussions.
As discussed in Attachment schedules for each of the four Quad Cities Unit 2 Dresden Unit 2 Quad Cities Unit 1 Dresden Unit 3 D, the earliest feasible units are as follows:
- April, 1988 Outage
- September, 1988 outage
- June, 1989 Outage
- December, 1989 outage implementation Although CECo believes'thfs is currently an achievable plan, it should be noted that the Quad Citie~'Unit' 2 schedule is based on an expedited modification approval and.the Dresden Unit 2 schedule is based on a tight material delive~y lead-time.
As suggested during tne Reference (5) confere~~e call, the remaining design activities in the modification process have:been adjusted to explicitly include a "Faiiure Modes and Effects Analysis".
This objective was previously accomplished as part of several other elements of the process rather th~n as a single., focused acti:vity.
Attachment D addresses this and several other* questions on the proposed modifications which the Staff raised in the conference call.
In addition to correcting the known swing bus design weakness, CECo has also contracted Sargent and Lundy Engineers to conduct a study of one of the four units (Quad Cities Unit 1) to assure that an assumed DC power failure would not adversely affect any other* safety equipment.
This study is scheduled for completion by June, 1988 including an evaluation of its results with respect to the remaining three units.
T. February 19, 1988 D.
REVISED LOCA ANALYSIS REPORT FOR QUAD CITIES lm Finally, enclosed as Attachment E is--the *revised SAFER/GESTR LOCA Analysis Report for Quad*Cities'(NEDC-31345P, ;Revision 1 dated January, 1988).
Changes to the report*are indicated by bars.in the.right margins.
The revision does not involve any change in the.'analysis or its results, but merely corrects the labeling of the cases pr'ev1ously referred to as "battery failure" to ref.lect the actual ECCS equipment failures (Diesel Generator and HPCI) which w9ul.d.result in the assumed remaining equipment (i.e., ADS, one core spray and two LPCis).
An unrelated and minor change is the updating of the report to include two additional* specific fuel bundle types of the previously approved GE8x8EB design which will be used in the next Unit 2 cycle.
The revisions to this report have been reviewed and approved by the CECo On-Site Review and Off-Site Review committees.
Please note that Revision 1 of the NEDC-31345P is considered Proprietary by General Electric on the same basis as the previous version of the report.
The affidavit provided previously by General Electric therefore remains applicable and is again included in Attachment E.
Please contact this office should further information be required.
Very truly yours,
~~
J. A.
Nuclear Licensing Administrator Attachments cc:
G.M. Holahan -
NRR A.B. Davis -
(Regional Admin.) RIII M. Grotenhuis - NRR T. Ross -
NRR NRC Resident Inspector - Dresden NRC Resident Inspector - Quad Cities M.C. Parker -
IONS D. Musolf - Northern states Power Company Cw/o Att. E)
- v. Crew - Iowa Electric Light and Power Company (w/o Att. E)
H. Pfefferlen - General Electric Company (w/o Att. E) 4208K
ATTACHMBNT A PROBABILITY OF COINCIDENT LOCA, LOOP AND BATTERY FAILURE GE's PRA analysis* uses data from WASH-1400 and estimates the probability of simultaneous LOCA, loss-of-offsite power, and loss of DC to be 2 x lo-12/yr for Monticello.
However, their analysis assumes a small break.
our scenario of concern involves a large break, which is a factor of ten less likely than a small break, but the factor for break size is not applicable.
Their analysis assumes redundant DC sources for the diesel-generator.
Not having that redundancy decreases DC reliability by a factor of ten.
Considering these factors yields a scenario estimate of 5.4 x 10 -12/yr, using their data and methodology.
- our analysis estimates the probability of simultaneous large LOCA, loss of off site power, and_.DC failure.
It is based on the Quad Cities Unit 1 specific design, and on failure data from IEEE-500-1984 and IEEE-500-1977.
The DC failures of interest are those which remove DC power from both the feed breaker to the swing MCC 18/19-5 and the Division II diesel generator.
Both are fed from the same*turbine building 125V DC r~serve bus lB-1, per Dwg.
4E-1318, Sh. 1, Rev. B.
According to Dwg. 4E-131.8B, Rev. c, the following failures can remove DC power to both loads during a loss of offsite power:
- 1)
Feed breaker* from t.b.. res. bus lB to t.b. res. bus lB-1 fails open.
- 2)
Fuse from 125V qc ma~n bus 2A. to t.b.* res. bus lB fails open.
- 3)
Fuse from battery bus #2 to main bus 2A fails open.
- 4)
Battery #2 fails.
Failures of the manual switches or the bus would be negligibl~ compared to the above failures.
IEEE-500 provides the following data for each of the above components:
Failure Com2onent
~
Mode source
~ Failure Rate
-6 Cir. Brkr.
(Composite)
All IEEE-500-1984 106
.23xl0 /hr.
-6 (Composite)
Open IEEE-500-1984 106
.OlxlO /hr.
w/o command
-6 Molded case All IEEE-500-1984 124 l.13xl0 /hr.
-6 Fuses Up to Open below IEEE-500-1977 193
.02lxl0 /hr.
l,OOOV rating Battery Secondary All IEEE-500-1984 81
-6 l.16xl0 /hr.
(Storage)
GE letter dated November 15, 1979 from A.A. Strod to G.H. Romer "Probability Assessment of a small Break Accident (SBA) with 125V DC Power Failure"
ATT. A 2 -
Molded case circuit breakers have higher f.ailure rates than the average for all breakers, but data was not provided for individual failure modes.
Assuming the same fraction are "open without command" for molded-case breakers as for all breakers, yields an estimated rate of "open without command" for molded case breakers of ~o5x10-6/hr. The data thus indicates that the failure probability of breakers -*and fuses is negligible compared to that of the batteries. Therefore, this analysis need only deal with battery failures.
The battery failure data in IEEE-500 cannot be taken at face value, however.
According to p.59 of IEEE-500-1984, 99.6% of the failures reported
- were discovered during testing. Therefore, since Quad Cities has a very good
. testing program, nearly all failures would be detected and corrected before an "accident".
Even then, the sorts of failures that would be undetected, otherwise, would be detected by the refueling-interval discharge test.
So, the mission time for these undetected failures is the average time since last discharge test, or 1.5 yr/2.
The expression for undetected failures is, therefore, Pu= (l.16xl0-6/hr)(l-99.6%)(1. 5 yr. x 365.25 days x24 hr) 2 yr day
= (l.l6x10-6/hr)(4xlo-3)(6.6xlo3 hr.)
Pu = 3.05 x 10-5/yr.
Detected failures would probably be uncovered by the DC bus undervoltage alarms.
Assume, for simplicity, that that alarm fails 1% of the time.
In that case, the mission time is the average time since last daily surveillance, or 24 hr/2.
Pa = (1.16 x lo-6/hr)(l x lo-2>c24hr) 2
= 13.9 x lo-8 = 1.4 x lo-1/yr.
This value is negligible compared to the undetected failure probability, above.
From NSAC-111, the generic loss-of-offsite power frequency is.078/yr.
This is certainly much higher than CECo. experience, for this number implies approximately one loss of offsite power during power operation every ten unit-years.
Yet, for this analysis, the generic value ~ill be used.
A standard large LOCA frequency is 1 x lo-4/yr. It dates back to WASH-1400, and it is certainly high, considering contemporary leak-before-break analysis.
Therefore, an upper-bound estimate of simultaneous loss-of-offsite power, LOCA, and DC failure is
ATT. A 3 -
The differences in numeric values between GE's analysis and ours is explained principally by different data.
Also, our DC analysis is more refined while their LOCA analysis includes factors for break location.
Despite these differences, both analyses show that this scenario is not significant to risk.
A rule-of-thumb cutoff for excluding events like aircraft crash from consideration for nuclear plants is lo-1 per plant year.
The above numbers are well below lo-1.
A typical plant core damage frequency from PRA's is lo-4.
Even if every simultaneous LOOP, LOCA, and DC failure resulted in core damage, the increase in core damage frequency would be negligible.
Therefore, the scenario under consideration is not a significant safety problem.
4208K
ATTACHMENT B DOCUMENTS RELATED TO ECCS SINGLE FAILURE ASSUMPTIONS INCLUDES:
- Discussion of Quad Cities design bases from the Attachment to Reference (l).
- List of relevant documents excerpted from the material provided for the Reference (5) conference call.
- Excerpts from the standard Review Plan, NUREG-0800, Rev. 2 - April 1984.
Similar statements were included in the previous revisions (1981 and 1975).
Please note that emphasis (underlining) has _been.added to I
highlight the sections of interest..
4208K
ATTACHMENT B QUAD CITIES STATION APPENDIX K FAILURE CRITERION In accordance with 10 CFR 50 Appendix K, loss of coolant accident analyses for Quad Cities Station are evaluated assuming that the worst single failure of ECCS equipment occurs.
A review of the Quad Cities design bases shows that these evaluations have always been performed assuming the failure of single active components, and not passive components such as batteries.
This assumption is documented in Licensing Topical Repor,t NED0-20566, "General Electric Company Analytical Model for Loss-of-Cool~nt Analysis tn Accordance with 10 CFR 50 Appendix K - Volume 1," (Reference )1*':).
The following excerpt from page I-302 of NED0-20566 clarifies the assumption made for QUad Cities station:
"General Electric \\ Compliance with I.D. l
- ,J The evaluation of the loss-of-coolant accident will be performed assuming the single active component failure that results.in the most severe consequences and assuming no credit for the station's normal auxiliary power.
The combination of ECC subsystems assumed to be operating shall be those remaining after the component failure has -
occurred.
The term "active component" means a component in which physical movement of moving parts must oc.cur in order to accomplish its safety function.
"Active component failure" mearis the failure of a component to accomplish its safety function due to failure of moving parts of the component.
A single failure may'be the failure of an individual component or it. may include the failures:()f.several c9mponents resulting from the cascading of the effect-of an initial fail,ure."
NED0-24146A (Reference 2, as Amended to provide MAPLHGR vs. exposure tables for new fuel types) is the plant specific loss-of-coolant accident analysis document which forms the licensing basis for QUad Cities Station prior to the application of SAFER/GESTR methodology.
It was not intended to change Quad Cities Station's licensing basis to consider the failure of passive components upon the application of SAFER/
GESTR methodology.
This consideration was discussed in References 3 and 4.
In Reference 3, section 2.2.1.1, it is clearly noted by General Electric that battery failure is not part of some plants' original design basis.
QUad Cities is one of.these early vintage BWR plants for which this is the case.
In section 2.2 of Reference 4 (NRC's Topical Report Evaluation of General
ATT. B 2 -
Electric SAFER/GESTR application methodology), it is stated that in determining limiting LOCA cases, "the scope of failures considered is the same as in the presently approved models."
This supports Commonwealth Edison's position that the failure of a 125 voe battery is a passive failure, and as such need not be considered in either the previous SAFE-REFLOOD or the new SAFER/GESTR QUad Cities Station LOCA analyses.*
REFERENCES l) NED0-20566 (Class I), "Ge~eral Elec'tric Company Analytical Model for Loss-of-Coolant Accident Analysis in Accordance With 10 CFR 50 Appendix K"
- Volume 1, dated January 1916.
- 2)
NED0-24146A (Class I), "Loss of Coolant Accident Analysis Report for Dresden Units 2,3 and QUad cities *units 1,2 Nuclear Power Stations," dated April 1919.
- 3)
NEDC-23185P (Class III), "The GESTR-LOCA and SAFER Models for the Evaluation of Loss-of-Coolant Accident," Volume III, dated OCtober 1984.
- 4)
Cecil o. Thomas (NRC) letter to J. F. Quick (GE), "Acceptance for Referencing of Licensing Topical Report NEDE-23185 Revision 1, Volume III (P)," dated June 1, 1984.
4208K
tt ATT. B DESIGN BASIS DOCUMENT SEARCH REFERENCES
- ORIGINAL DESIGN INTENT - WORST SINGLE DC SYSTEM FAILURE WOULD RESULT IN LOSS OF 1/2 OF THE CORE COOLING EQUIPMENT
- 1.
DRESDEN FSAR AMENDMENT 11/12, QUESTION.II.B.11-1
- 2.
JANECEK TO. IPPOLITO LETTER DATED JUNE 12, 1980
- 3. QUAD-CITIES UFSAR SECTION 6.2.7.6
- SINGLE ACTIVE FAILURE BASIS FOR LOCA ANALYSES
- 1.
FSAR APPENDIX B-22, ENGINEERED SAFETY-FEATURES
- 2. **G.E. ECCS REPORT NEDE-20566, SECTION 1.D.l, JANUARY 1976
- 3.
NRC ACCEPTANCE LETTER FROM TEDESCO TO SHERWOOD DATED FEBRUARY 1981
- 4.
SAFER/GESTR NEDC-23785P* VOLUME III DATED OCTOBER 1984.
- 5.
NRC ACCEPTANCE OF NEDE-23785 VOLUME III DATED JUNE 1984
- NOT PREVIOUSLY AWARE OF LPCI* INJECTION VALVE FAILURE DUE TO LOSS OF BATTERY AND LOSS OF OFFSITE POWER
- 1.
DRESDEN FSAR AMENDMENT 11/12 *
- 2.
JANECEK TO IPPOLITO LETTER DATED JUNE 12, 1980.
- 3.
NUREG-0813 FOR DRESDEN UNIT 2 SEP, SECTION 4.23, PAGES 4-30 *.
JSA 2-1-88
e,:ATT. B (NUREG-0800 Exe.ts) continued effective core cooling, to permit appropriate periodic inspec-tion of important components, and to permit appropriate periodic pressure and functional testing.
G.
10 CFR Part 50, §50.46, and Appendix K*to 10 CFR Part 50 as it relates to the ECCS being designed so that its cooling performance is in accordance with an acceptable evaluation model.
Specific acceptance criteria, Regulatory Guides, and Task Action Plan items that provide information, recommendatiqns, and guidance and in general describe a basis acceptable to the staff that may be used to implement the requirements of the Commission regulations identified above are as follows:
In regard to the ECCS acceptance criteria (Ref. 1), the five major performance criteria deal with:
- 1.
Peak cl~dding temperature.
- 2.
Maximum calculated cladding oxidation.
- 3.
Maximum hydrogen generation.
- 4.
Coolable core geometry.
- 5.
Long-term cooling.
These areas are reviewed as a part of the effort associated with the LOCA analysis (SRP Section 15.£.5).
However, the impact of various postulated single failures on the operability of the ECCS is evaluated under this SRP section.
'The ECCS must meet the requirements of GDC 35 (Ref. 6).
The system must have alternate sources of electric power, as required by GDC 17 (Ref. 4), and must be able to withstand a single failure.
The ECCS should retain its capability to cool the core in the event of a failure of any single active component dur-ing the short term immediately following an accident, or a single active or passive failure during the long-term recirculation cooling phase following an accident.
The ECCS must be designed to permit periodic in~ervice inspection of important components, such as spray rings in the reactor pressure vessel, water injection nozzles, piping, pumps, and valves in accordance with the requirements of GDC 36 (Ref. 7).
The ECCS must be designed to permit testing of the operability of the system throughout the life of the plant, including the full operational sequence that brings *the system into operation, as required by GDC 37 (Ref. 8).
The combined reactivity control system capability associated with ECCS must meet the requirements of GDC 27 (Ref. 5) and should conform to the recommenda-tion of Regulatory Guide 1.47 (Ref. 11).
The primary mode of actuation for the ECCS must be automatic, and actuation must be initiated by signals of suitable diversity *and redundance.
Provisions should also be made for manual actuation,
.. monitoring, and control of the ECCS from the reactor control room.
The design of the ECCS should conform to the recommendations of Regulatory Guide 1.1 (Ref. 9).
Design features and operating procedures, designed to prevent damaging water hammer due to such mechanisms as voided discharge lines and water entrainment 6.3-4 Rev. 2 -
April 1984 *
~TT.. B (NUREG-0800 Excerpe
- 21.
The RSB reviewer consults with the ICSB reviewer to:
- a.
Confirm that the power requirements of the ECCS, including the timing of electrical loads, are compatible with the design of onsite emergency pow~r systems, both a-c and d-c.
- b.
Confirm that there are sufficient instrumentation and controls avail-able to the reactor operator to provide adequate information in the control room to assist in assessing post-LOCA conditions, including the more significant parameters such as coolant flow, coolant temperature, and containment pressure.
If ECCS flow is diverted as a backup to other safeguards systems, the reviewer confirms that instrumentation and controls are available to provide sufficient information in the control room to determine that adequate core cooling is being provided.
- c.
Confirm that automatic actuation and remote-manual valve controls are capable of performing the functions required, that suitable interlocks are provided, which do not impair separation of power trains or inhibit the required valve motions, and that instrumenta-tion and controls have sufficient redundancy to satisfy the single failur@ criterion.
- 22.
Analyses are provided by the applicant in Chapter 15 of the SAR to assess the capability of the ECCS to meet functional requirements.
These analyses are reviewed by the RSB, as described in SRP Section 15.6.5, to determine conformance to the acceptance criteria. for ECCS.
However, the following portions of the review of ECCS response in loss-of-coolant accidents are performed by the RSB reviewer under this SRP section:
- a.
The lower limit of break size for which ECCS operation is required is established; i.e., the maximum break size for which normal reactor coolant makeup systems can maintain reactor pressure and coolant level is determined.
The capability of the ECCS to actuate and perform at this lower limit of break size is confirmed.
- b.
The reviewer confirms that the analyses take into account a variety of potential locations for postulated pipe breaks, including ECCS -
injection lines.
- c.
The reviewer confirms that the analyses take into account a variety of single active failures.
The reviewer should keep in mind that different single failures may be limiting, depending on the particular break location and break size postulated.
- d.
The ECCS component response times (e.g., for valves, pumps, power supply) are reviewed to confirm that they are within the delay times used in the accident analyses.
- e.
The ECCS design adequacy for all modes of reactor operation (e.g.,
full power, low power, hot standby, cold shutdown, partial loop isolation) is confirmed.
- 23.
The proposed plant technical specifications are reviewed to:
6.3-9 Rev. 2 - April 1984